Abstract
In environmental statistics, surveys on the structure of biological communities are generally carried out by focusing on diversity indexes. A more complete analysis may be performed by means of an appropriate function giving a spectrum of different measures of diversity: diversity profiles. They can be expressed as a function of the unknown abundance vector of the ecological population. In this paper we develop a non parametric approach based on bootstrap in order to make inference on diversity profiles. The proposed procedure is applied on biological data of four parks in Milan, Italy.
Highlights
In many environmental and ecological surveys the aim is to evaluate the diversity of biological communities
In this paper we develop a non parametric approach based on bootstrap in order to make inference on diversity profiles
We suggest the first-order Taylor expansion of (1) in order to achieve the β diversity profile as a linear combination of the entries of the unknown relative abundance vector p followed by a non parametric methodology based on bootstrap which tests the presence of an intrinsic diversity ordering between two communities under study
Summary
In many environmental and ecological surveys the aim is to evaluate the diversity of biological communities. We give a solution to these problems by suitably using a nonparametric technique (Beran, 1988) adopted to build simultaneous confidence sets of a family of parametric functions This procedure requires the estimator to be a linear function of the components of the mean vector in order to have reasonable asymptotically properties. At this purpose, we suggest the first-order Taylor expansion of (1) in order to achieve the β diversity profile as a linear combination of the entries of the unknown relative abundance vector p followed by a non parametric methodology based on bootstrap which tests the presence of an intrinsic diversity ordering between two communities under study. The developed procedure is applied to a real data set
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
